At Voxel Rx we are developing the next generation of deep learning tools to better understand Alzheimer's Disease. Through this understanding, we aim to discover biomarkers for the diagnosis and early detection of neurological dysfunction so that treatment and intervention protocols can be established in this decade. Exponential problems require ideas that scale. Our tools leverage the latest cloud technologies including distributed GPU processing, as well as unsupervised feature learning and hierarchical forms of representation, that are capable of learning the discriminative, latent structure in data that is potentially unavailable to human physicians.
Alzheimer's disease (AD) is becoming increasingly prevalent in a society where we live longer than before with estimates of 130+ million worldwide affected by this one neurodegenerative disease alone. AD not only impacts those inflicted by the disease, but also the loved ones surrounding them. The problem is that we still don't understand the disease and therefore the early detection of AD onset is not possible for intervention before the disease progresses too far.
Our solution to this problem is to develop deep learning algorithms that can discover the latent structure in the neuroimaging data that can allow us to diagnose and detect the early onset of AD. More specifically, we use a combination of unsupervised and supervised learning techniques in deep neural networks to learn diagnostic biomarkers and transform volumetric sMRI data into actionable feedback.
The solution we provide leverages exponential trends in hardware development of Graphical Processing Units for training and implementing neural networks, increasing neuroimaging resolutions and screening ubiquity to build more complex and generalizable models, and basic research on deep learning that continues to push the state of the art.
Many existing health records are still physical copies and need to be transferred to electronic health records to keep up with the pace of health care. We are developing machine learning solutions to transfer scanned images of documents to text with labels so that they can be stored as EHR. With this transition, massive data mining can be performed for classifying patient types, predicting patient ailments and predicting intervention outcomes.
William Edward Hahn co-founded the Machine Perception and Cognitive Robotics Laboratory in 2014 with the purpose of building a team of creative scientists, programmers, and engineers to research perception, computation, and robotics. His PhD thesis explores the intersection of computational neuroscience, non-linear signal processing, and computer vision.
Website
Machine Learning and Cognitive Robotics
Bryan is currently a PhD student at the Center for Complex Systems and Brain Sciences in the Cognitive Neurodynamics laboratory. He holds a bachelor's degree in math, bachelor's degree in philosophy and master's degree in computer science. Bryan is also an accomplished entrepreneur with 8 years of C-suite experience in the healthcare industry where he grew and helped sell a $13 million home care agency. He also ran a design and engineering firm and a tech startup focused on custom app and website development. Bryan is a founding board member and chair of the startup council at Palm Beach Tech, a nonprofit membership association focused on uniting the technology industry in Palm Beach County. He is part of the entrepreneur task force of the Business Development Board of Palm Beach County and serves as Treasurer of his HOA's Board of Directors.
Personal Blog
Twitter
Please feel free to drop us a line. We would love to hear from you.
- Voxel Rx Team